Automatic Stabilizers and Economic Crisis: US vs. EuropeMathias Dolls�, Clemens Fuestz, Andreas Peichlx

This version: January 27, 2010{

Abstract: This paper analyzes the e¤ectiveness of the tax and transfer systemsin the European Union and the US to act as an automatic stabilizer in the current

economic crisis. We �nd that automatic stabilizers absorb 38 per cent of a propor-

tional income shock in the EU, compared to 32 per cent in the US. In the case of

an unemployment shock 48 per cent of the shock are absorbed in the EU, compared

to 34 per cent in the US. This cushioning of disposable income leads to a demand

stabilization of 26 to 35 per cent in the EU and 19 per cent in the US. There is

large heterogeneity within the EU. Automatic stabilizers in Eastern and Southern

Europe are much lower than in Central and Northern European countries. We also

investigate whether countries with weak automatic stabilizers have enacted larger

�scal stimulus programs. We �nd no evidence supporting this view. However, we

�nd that active �scal policy is lower in more open economies.

�CGS - University of Cologne and IZA, dolls@wiso.uni-koeln.dezUniversity of Oxford (Centre for Business Taxation), University of Cologne, CESifo and IZA,

clemens.fuest@sbs.ox.ac.ukxIZA Bonn, University of Cologne and ISER, peichl@iza.org{This paper uses EUROMOD version D21 and TAXSIM v9. EUROMOD and TAXSIM are continually being

improved and updated and the results presented here represent the best available at the time of writing. Our version

of TAXSIM is based on the Survey of Consumer Finances (SCF) by the Federal Reserve Board. EUROMOD relies

on micro-data from 17 di¤erent sources for 19 countries. These are the ECHP and EU-SILC by Eurostat, the

Austrian version of the ECHP by Statistik Austria; the PSBH by the University of Liège and the University of

Antwerp; the Estonian HBS by Statistics Estonia; the Income Distribution Survey by Statistics Finland; the EBF

by INSEE; the GSOEP by DIW Berlin; the Greek HBS by the National Statistical Service of Greece; the Living in

Ireland Survey by the Economic and Social Research Institute; the SHIW by the Bank of Italy; the PSELL-2 by

CEPS/INSTEAD; the SEP by Statistics Netherlands; the Polish HBS by Warsaw University; the Slovenian HBS

and Personal Income Tax database by the Statistical O¢ ce of Slovenia; the Income Distribution Survey by Statistics

Sweden; and the FES by the UK O¢ ce for National Statistics (ONS) through the Data Archive. Material from the

FES is Crown Copyright and is used by permission. Neither the ONS nor the Data Archive bears any responsibility

for the analysis or interpretation of the data reported here. An equivalent disclaimer applies for all other data

sources and their respective providers.

This paper is partly based on work carried out during Andreas Peichl�s visit to the European Centre for Analysis

in the Social Sciences (ECASS) at the Institute for Social and Economic Research (ISER), University of Essex,

supported by the Access to Research Infrastructures action under the EU Improving Human Potential Programme.

Andreas Peichl is grateful for �nancial support by Deutsche Forschungsgemeinschaft DFG (PE1675). We would like

to thank Danny Blanch�ower, Dean Baker, Horacio Levy, Tor�nn Harding, participants of the 2009 IZA/CEPR

ESSLE and IZA Prize conferences and the 6th German-Norwegian Seminar on Public Economics (CESifo) as

well as seminar participants in Bonn, Cologne, Nuremberg, Siegen and at the Worldbank for helpful comments

and suggestions. We are grateful to Daniel Feenberg for granting us access to TAXSIM and helping us with our

simulations. We are indebted to all past and current members of the EUROMOD consortium for the construction

and development of EUROMOD. The usual disclaimer applies.

JEL Codes: E32, E63, H2, H31

Keywords: Automatic Stabilization, Crisis, Liquidity Constraints, Fiscal Stimulus

I

1 Introduction

In the current economic crisis, the workings of automatic stabilizers are widely seen

to play a key role in stabilizing demand and output. Automatic stabilizers are usu-

ally de�ned as those elements of �scal policy which mitigate output �uctuations

without discretionary government action (see, e.g., Eaton and Rosen (1980)). De-

spite the importance of automatic stabilizers for stabilizing the economy, "very little

work has been done on automatic stabilization [...] in the last 20 years" (Blanchard

(2006)). However, especially in the current crisis, it is important to assess the con-

tribution of automatic stabilizers to overall �scal expansion and to compare their

magnitude across countries. Previous research on automatic stabilization has mainly

relied on macro data. Exceptions based on micro data are Auerbach and Feenberg

(2000) for the US and Mabbett and Schelkle (2007) for the EU-15. Much more

comparative work based on micro data has been conducted on the di¤erences in the

tax wedge and e¤ective marginal tax rates between the US and European countries

(see, e.g., Piketty and Saez (2007)).

In this paper, we combine these two strands of the literature to compare the

magnitude and composition of automatic stabilization between the US and Europe

based on micro data estimates. We analyze the impact of automatic stabilizers

using microsimulation models for 19 European countries (EUROMOD) and the US

(TAXSIM). The microsimulation approach allows us to investigate the causal e¤ects

of di¤erent types of shocks on household disposable income, holding everything

else constant and therefore avoiding endogeneity problems (see Bourguignon and

Spadaro (2006)). We can hence single out the role of automatic stabilization which

is not possible in an ex-post evaluation (or with macro data) as it is not possible

to disentangle the e¤ects of automatic stabilizers, active �scal and monetary policy

and behavioral responses like changes in labor supply or disability bene�t take-up.

We run two controlled experiments of macro shocks to income and employment.

The �rst is a proportional decline in household gross income by 5% (income shock).

This is the usual way of modeling shocks in simulation studies analyzing automatic

stabilizers. However, economic downturns typically a¤ect households asymmetri-

cally, with some households losing their jobs and su¤ering a sharp decline in income

and other households being much less a¤ected, as wages are usually rigid in the

short term. We therefore consider a second macro shock where some households

become unemployed, so that the unemployment rate increases by �ve percentage

1

points (unemployment shock). We show that these two types of shocks and the re-

sulting stabilization coe¢ cients can be interpreted as an average e¤ective marginal

tax rate (EMTR) for the whole tax bene�t system at the intensive (proportional

income shock) or extensive (unemployment shock) margin. After identifying the

e¤ects of these shocks on disposable income, we use methods developed by Zeldes

(1989) to estimate the prevalence of credit constraints among households. On this

basis, we calculate how the stabilization of disposable income can translate into

demand stabilization.

As our measure of automatic stabilization, we extend the normalized tax change

(Auerbach and Feenberg (2000)) to include other taxes as well as social contribu-

tions and bene�ts. Our income stabilization coe¢ cient relates the shock absorption

of the whole tax and transfer system to the overall size of the income shock. We take

into account personal income taxes (at all government levels), social insurance con-

tributions and payroll taxes paid by employers and employees, value added or sales

taxes as well as transfers to private households like, e.g., unemployment bene�ts.1

Computations are done according to the tax bene�t rules which were in force before

2008 in order to avoid an endogeneity problem resulting from any policy responses

after the start of the crisis.

What does the present paper contribute to the literature? First, previous studies

have focused on proportional income shocks whereas our analysis shows that auto-

matic stabilizers work very di¤erently in the case of unemployment shocks, which

a¤ect households asymmetrically.2 This is especially important for assessing the ef-

fectiveness of automatic stabilizers in the current economic crisis. Second, we extend

the micro data measure on automatic stabilization to di¤erent taxes and bene�ts.

Our analysis includes a decomposition of the overall stabilization e¤ects into the

contributions of taxes, social insurance contributions and bene�ts. A further dif-

ference between our study and Auerbach and Feenberg (2000) is that we take into

account unemployment bene�ts and state level income taxes as well as consumption

taxes. This explains why our estimates of overall automatic stabilization e¤ects in

the US are higher. Third, to the best of our knowledge, our study is the �rst to

1We abstract from other taxes, in particular corporate income taxes. For an analysis of auto-matic stabilizers in the corporate tax system see Devereux and Fuest (forthcoming) and Buettnerand Fuest (2009).

2Auerbach and Feenberg (2000) do consider a shock where households at di¤erent incomelevels are a¤ected di¤erently, but the results are very similar to the case of a symmetric shock.Our analysis con�rms this for the US, but not for Europe.

2

estimate the prevalence of liquidity constraints for EU household data.3 This is of

key importance for assessing the role of automatic stabilizers for demand smoothing.

Fourth, we extend the analysis to more recent years and countries - including tran-

sition countries from Eastern Europe - and we compare the US and Europe within

the same microeconometric framework. Finally, we shed light on the issue whether

macro indicators are a good proxy for micro data based stabilization coe¢ cients.

We also investigate whether bigger governments or more open economies lead to

higher / lower automatic stabilizers.

We show that our extensions to previous research are important for the compar-

ison between the U.S. and Europe as they help to identify driving forces in auto-

matic stabilization. Our analysis leads to the following main results. In the case of

an income shock, approximately 38% of the shock would be absorbed by automatic

stabilizers in the EU. For the US, we �nd a value of 32%. This is surprising because

automatic stabilizers in Europe are usually perceived to be drastically higher than

in the US. Our results qualify this view to some extent, at least as far as propor-

tional shocks on household income are concerned. When looking at the personal

income tax only, the values for the US are even higher than the EU average. Within

the EU, there is considerable heterogeneity, and results for overall stabilization of

disposable income range from a value of 25% for Estonia to 56% for Denmark. In

general automatic stabilizers in Eastern and Southern European countries are con-

siderably lower than in Continental and Northern European countries. In the case

of the unemployment shock, the di¤erence between the EU and the US is larger.

EU automatic stabilizers absorb 48% of the shock whereas the stabilization e¤ect in

the US is only 34%. Again, there is considerable heterogeneity within the EU.

How does this cushioning of shocks translate into demand stabilization? Since

demand stabilization can only be achieved for liquidity constrained households, the

picture changes signi�cantly. For the income shock, the cushioning e¤ect of auto-

matic stabilizers is now equal to 26% in the EU. For the US, we �nd a value of

19%, which is again rather similar. For the unemployment shock, however, we �nd

a large di¤erence. In the EU, the stabilization e¤ect is equal to 35% whereas the

value for the US (19%) is similar to the value for the income shock. These results

suggest that social transfers, in particular the rather generous systems of unemploy-

3There are several studies on liquidity constraints and the responsiveness of households to taxchanges for the US (see, e.g., Zeldes (1989), Parker (1999), Souleles (1999), Johnson et al. (2006),Shapiro and Slemrod (1995, 2003, 2009))

3

ment insurance in Europe, play a key role for demand stabilization and explain an

important part of the di¤erence in automatic stabilizers between Europe and the

US.

A �nal issue we discuss in the paper is how �scal stimulus programs of individual

countries are related to automatic stabilizers. In particular, we ask whether coun-

tries with low automatic stabilizers have tried to compensate this by larger �scal

stimuli, but we �nd no correlation between the size of �scal stimulus programs and

automatic stabilizers. However, we �nd that active �scal policy is lower in more

open economies.

The paper is structured as follows. In Section 2 we provide a short overview

on previous research with respect to automatic stabilization and comparisons of US

and European tax bene�t systems. In addition, we discuss how stabilization e¤ects

can be measured. Section 3 describes the microsimulation models EUROMOD and

TAXSIM and the di¤erent macro shock scenarios we consider. Section 4 presents

the results on automatic stabilization which are discussed in Section 5 together with

potential limitations of our approach. In Section 6, we shortly discuss distributional

implications of the analyzed macro shock scenarios. Section 7 concludes.

2 Previous research and theoretical framework

2.1 Previous research

There are two strands of literature which are related to our paper. First, in the

literature on the analysis and measurement of automatic �scal stabilizers, scholars

have estimated automatic stabilizers mostly based on macro data. Related to this

literature are studies which have investigated the relationship between openness,

output volatility and government size, a proxy for automatic �scal stabilizer. Second,

there is a growing literature on international comparisons of income tax systems.

This literature has mainly relied on micro data and the simulation approach in order

to take into account the heterogeneity of the population.

In the empirical literature on automatic stabilizers4, two types of studies prevail:

macro data studies and micro data estimates.5 The common baseline of macro data4A theoretical analysis of automatic stabilizers in a real business cycle model can be found in

Galí (1994).5Early estimates on the responsiveness of the tax system to income �uctuations are discussed

in the Appendix of Goode (1976). More recent contributions include Fatàs and Mihov (2001),

4

studies is to measure the cyclical elasticitiy of di¤erent budget components such

as the income tax, social security contributions, the corporate tax, indirect taxes

or unemployment bene�ts. Di¤erent approaches have been proposed, for exam-

ple regressing changes in �scal variables on the growth rate of GDP or estimating

elasticities on the basis of macro-econometric models.6 Sachs and Sala-i Martin

(1992) and Bayoumi and Masson (1995) use time series data and �nd values of

30%-40% for disposable income stabilization in the US. However, these approaches

face several drawbacks from which the most serious one is probably the impossi-

bility of separating discretionary actions from automatic stabilizers in combination

with identi�cation problems resulting from endogenous regressors. Related to the

literature on macro estimations of automatic stabilization are studies that focus on

the relationship between output volatility, public sector size and openness of the

economy (Cameron (1978), Galí (1994), Rodrik (1998), Fatàs and Mihov (2001),

Auerbach and Hassett (2002)).

Much less work has been done on the exact identi�cation of automatic stabilizers

resulting from tax and transfer systems with micro data. Auerbach and Feenberg

(2000) use the NBER�s microsimulation model TAXSIM to estimate the automatic

stabilization for the US from 1962-95 and �nd values for the stabilization of dispos-

able income ranging between 25%-35%. Auerbach (2009) has updated this analysis

and �nds a value around 25% for more recent years. Mabbett and Schelkle (2007)

conduct a similar analysis for 15 Western European countries in 1998 and �nd higher

stabilization e¤ects than in the US, with results ranging from 32%-58%.7 How does

this smoothing of disposable income a¤ect household demand? To the best of our

knowledge, Auerbach and Feenberg (2000) is the only simulation study which tries

to estimate the demand e¤ect taking into account liquidity constraints. They use

the method suggested by Zeldes (1989) and �nd that approximately two thirds of

all households are likely to be liquidity constrained. Given this, the contribution of

automatic stabilizers to demand smoothing is reduced to approximately 15% of the

initial income shock.

The second strand this paper adds to is the literature on international compar-

Blanchard and Perotti (2002), Mélitz and Zumer (2002).6Cf. van den Noord (2000) or Girouard and André (2005).7Mabbett and Schelkle (2007) rely for their analysis (which is a more recent version of Mabbett

(2004)) on the results from an in�ation scenario taken from Immvervoll et al. (2006) who use themicrosimulation model EUROMOD to increase earnings by 10% in order to simulate the sensitivityof poverty indicators with respect to macro level changes.

5

isons of income tax systems in terms of e¤ective average and marginal tax rates,

and individual tax wedges between the US and European countries. Here, micro

data allow much more detailed analyses which are not possible with macro data.

Piketty and Saez (2007) use large public micro-�le tax return data for the US to

compute average tax rates for �ve federal taxes and di¤erent income groups and

complement the analysis for the US with a comparison to France and the UK. A

key �nding from their analysis is that today (and in contrast to 1970), France as

a continental European welfare state has higher average tax rates compared with

the two Anglo-Saxon countries and its tax system possesses a higher degree of pro-

gressivity. Immvervoll (2004) discusses conceptual issues with regard to macro- and

micro-based measures of the tax burden and compares e¤ective tax rates in four-

teen EU Member States. In general, he �nds a large heterogeneity across countries

with average and marginal e¤ective tax rates lowest in southern European coun-

tries. Other studies take as given that European tax systems reveal a higher degree

of progressivity (e.g. Alesina and Glaeser (2004)) or higher (marginal) tax rates in

general (e.g. Prescott (2004) or Alesina et al. (2005)) and discuss if and to what

extent di¤erences in economic outcomes such as hours worked can be explained by

di¤erent tax structures. This paper providing new measures of the average e¤ective

marginal tax rate (EMTR) both at the intensive and extensive margin for the US

and 19 European countries shall help to shed further light on existing di¤erences

between the US and European tax and transfer systems.

2.2 Theoretical framework

The extent to which automatic stabilizers mitigate the impact of income shocks on

household demand essentially depends on two factors. Firstly, the tax and transfer

system determines the way in which a given shock to gross income translates into a

change in disposable income. For instance, in the presence of a proportional income

tax with a tax rate of 40%, a shock on gross income of one hundred Euros leads to

a decline in disposable income of 60 Euros. In this case, the tax absorbs 40% of the

shock to gross income. A progressive tax, in turn, would have a stronger stabilizing

e¤ect. Secondly, the link between current disposable income and current demand

for goods and services is crucial. If the income shock is perceived as transitory and

current demand depends on some concept of permanent income, and if households

can borrow, their demand will not change. In this case, the impact of automatic

6

stabilizers on current demand would be equal to zero. Things are di¤erent, though,

if households are liquidity constrained. In this case, their current expenditures do

depend on disposable income so that automatic stabilizers play a role.

A common measure for estimating automatic stabilization is the "normalized tax

change" used by Auerbach and Feenberg (2000) which can be interpreted as "the

tax system�s built-in �exibility" (Pechman (1973, 1987)). It shows how changes

in market income translate into changes in disposable income through changes in

personal income tax payments.

Market income Y Mi of individual i is de�ned as the sum of all incomes from

market activities:

Y Mi = Ei +Qi + Ii + Pi +Oi (1)

where Ei are earnings, Qi business income, Ii capital income, Pi property income,

and Oi other income. Disposable income Y Di is de�ned as market income minus net

government intervention Gi = Ti + Si �Bi :

Y Di = Y Mi �Gi = Y Mi � (Ti + Si �Bi) (2)

where Ti are direct taxes, Si employee social insurance contributions, and Bi are

social cash bene�ts (i.e. negative taxes). Note that an extended analysis includ-

ing employer social insurance contributions and consumption taxes is presented in

Section 4.3.

We extend the "normalized tax change" to include other taxes as well as social

insurance contributions and bene�ts. This allows us to relate the shock absorption

by the whole tax and transfer system to the overall size of the income shock. We

take into account personal income taxes (at all government levels), social insurance

contributions as well as payroll taxes and transfers to private households like, e.g.,

unemployment bene�ts. In the following, we simply refer to our measure as the in-

come stabilization coe¢ cient � I . We derive � I from a general functional relationship

between disposable income and market income:

� I = � I(Y M ; T; S;B): (3)

The derivation can be either done on the macro or on the micro level. On

the macro level, it holds that the aggregate change in market income (�Y M) is

7

transmitted via � I into an aggregate change in disposable income (�Y D):

�Y D = (1� �)�Y M (4)

It can be easily shown that � I stabilizes GDP:

GDP = C + I +G+NX

�GDP = �C +�I +�G+�NX

�GDP = (1� �)�Y M +�G+�NX (5)

with C private consumption, I = S private investments (savings), G government

activity and NX net exports, and Y D = C + S.

However, one problem when comupting � I with macro data is that this data in-

cludes behavioral and general equilibrium e¤ects as well as active policy. Therefore,

a measure of automatic stabilization based on macro data captures all these e¤ects.

In order to single out the pure size of automatic stabilization, we compute � I using

arithmetic changes (�) in total disposable income (P

i�YDi ) and market income

(P

i�YMi ) based on micro level information:

Xi

�Y Di = (1� � I)Xi

�Y Mi

� I = 1�P

i�YDiP

i�YMi

=

Pi

��Y Mi ��Y Di

�Pi�Y

Mi

=

Pi�GiPi�Y

Mi

(6)

� I measures the sensitivity of disposable income, Y Di ; with respect to market

income, Y Mi . The higher �I , the stronger the stabilization e¤ect, e.g. � I = 0:4 im-

plies that 40% of the income shock is absorbed by the tax bene�t system. Note that

the income stabilization coe¢ cient is not only determined by the size of government

(e.g. measured as expenditure or revenue in percent of GDP) but also depends on

the structure of the tax bene�t system and the design of the di¤erent components.

Note that the de�nition of � I resembles that of an average e¤ective marginal tax

rate (EMTR), which is usually computed in this way using micro data (Immvervoll

(2004)). In the case of the proportional income shock, � I can be interpreted as the

EMTR along the intensive margin, whereas in the case of the unemployment shock,

8

it resembles the EMTR along the extensive margin (participation tax rate, see, e.g.,

Saez (2001, 2002), Kleven and Kreiner (2006) or Immervoll et al. (2007)).

Furthermore, it is important to explore the extent to which di¤erent individual

components of the tax transfer system contribute to stabilization. Comparing tax

bene�t systems in Europe and the US, we are interested in the weight of each

component in the respective country. We therefore decompose the coe¢ cient into

its components which include taxes, social insurance contributions and bene�ts:

� I =Xf

� If = �IT+�

IS+�

IB =

Pi�TiPi�Y

Mi

+

Pi�SiPi�Y

Mi

�P

i�BiPi�Y

Mi

=

Pi (�Ti +�Si ��Bi)P

i�YMi

(7)

However, in order to stabilize �nal demand and output, the cushioning e¤ect on

disposable income has to be transmitted to expenditures for goods and services. If

current demand depends on some concept of permanent income, demand will not

change in response to a transitory income shock. Things are di¤erent, though, if

households are liquidity constrained and cannot borrow. In this case, their current

expenditures do depend on disposable income so that automatic stabilizers play a

role. Following Auerbach and Feenberg (2000), we assume that households who

face liquidity constraints fully adjust consumption expenditure after changes in dis-

posable income while no such behavior occurs among households without liquidity

constraints.8 The adjustment of liquidity constrained households takes place such

that changes in disposable income are equal to changes in consumption. Hence, the

coe¢ cient which measures stabilization of aggregate demand becomes:

�C = 1�P

i�CLQiP

i�YMi

(8)

where �CLQi denotes the consumption response of liquidity constrained house-

holds. In the following, we refer to �C as the demand stabilization coe¢ cient. In

order to explore the sensitivity of our estimates of the demand stabilization coe¢ -

cient with respect to the way in which liquidity constrained households are identi�ed,

we choose two di¤erent approaches. In the �rst one, we use the same approach as

8Note that the term "liquidity constraint" does not have to be interpreted in an absoluteinability to borrow but can also come in a milder form of a substantial di¤erence between borrowingand lending rates which can result in distortions of the timing of purchases. Note further that ourdemand stabilization coe¢ cient does not predict the overall change of �nal demand, but the extentto which demand is stabilized by the tax bene�t system.

9

Auerbach and Feenberg (2000) and follow Zeldes (1989) to split the samples accord-

ing to a speci�c wealth to income ratio. A household is liquidity constrained if its

capitalized wealth Wi is less than the disposable income of at least two months, i.e:

LQi = 1

�Wi �

2

12Y Di

�(9)

The second approach simply considers the bottom 75% of the gross income dis-

tribution to be liquidity constrained.

2.3 Extension

In this section, we include employer social insurance contribution and consumption

taxes in our analysis. The inclusion of employer social insurance contributions is

of interest for two reasons. First, in our basic analysis, gross income is identical

to market income by de�nition. This does not hold, however, if employer social

insurance contributions are included. In this case, gross income is de�ned as follows:

Y Gi = Y Mi + SERi (10)

where Y Gi is gross income, Y Mi market income and SERi employer social insur-

ance contributions. The fact that gross income is larger than market income when

employer SIC are included explains why stabilization results may not only di¤er for

SIC, but also for taxes and bene�ts. Second, legislation with respect to the �nanc-

ing of social insurance can be very di¤erent across countries.9 Contributions are not

shared equally between employers and employees in all countries as it is the case,

e.g., in the US or Germany.

The income stabilization coe¢ cient can now be decomposed as follows:

� I =Xf

� If =

Pi (�Ti +�Si ��Bi +�CTi)P

i�YMi

(11)

where �Si includes the change in both employer and employee social insurance

contributions and �CTi is the di¤erence in consumption tax payments given that

all disposable income is consumed. In our computations, we use the implicit tax

9E.g. Ankrom (1993) also follows this approach. He argues that ignoring employers� socialcontributions would implicitly assume that the wage elasticity of labour supply is in�nite. However,empirical analyses rather suggest the opposite and therefore this can signi�cantly bias the resultsof the e¤ectiveness of tax and transfer systems.

10

rate (ITR) on consumption taken from European Commission (2009b) for European

countries and McIntyre et al. (2003) for the US, which is a measure for the e¤ective

tax burden as it includes several consumption taxes such as VAT, energy and other

excise taxes.10

3 Data and methodology

3.1 Microsimulation using TAXSIM and EUROMOD

We use microsimulation techniques to simulate taxes, bene�ts and disposable in-

come under di¤erent scenarios for a representative micro-data sample of households.

Simulation analysis allows conducting a controlled experiment by changing the pa-

rameters of interest while holding everything else constant (cf. Bourguignon and

Spadaro (2006)). We therefore do not have to deal with endogeneity problems when

identifying the e¤ects of the policy reform under consideration.

Simulations are carried out using TAXSIM - the NBER�s microsimulation model

for calculating liabilities under US Federal and State income tax laws from individual

data - and EUROMOD, a static tax-bene�t model for 19 EU countries, which was

designed for comparative analysis.11 The models can simulate most direct taxes and

bene�ts except those based on previous contributions as this information is usually

not available from the cross-sectional survey data used as input datasets. Informa-

tion on these instruments is taken directly from the original data sources. Both

models assume full bene�t take-up and tax compliance, focusing on the intended

e¤ects of tax-bene�t systems. The main stages of the simulations are the following.

First, a micro-data sample and tax-bene�t rules are read into the model. Then for

each tax and bene�t instrument, the model constructs corresponding assessment

10Note that this approach implicitly assumes that the marginal savings rate remains unchangedbefore and after the shock. This does not have to be the case as some households might have todecrease their savings due to decreasing disposable income, whereas others might even increase it(precautionary savings).

11For more information on TAXSIM see Feenberg and Coutts (1993) or visithttp://www.nber.org/taxsim/. For further information on EUROMOD see Sutherland(2001, 2007). There are also country reports available with detailed informationon the input data, the modeling and validation of each tax bene�t system, seehttp://www.iser.essex.ac.uk/research/euromod. The tax-bene�t systems included in the modelhave been validated against aggregated administrative statistics as well as national tax-bene�tmodels (where available), and the robustness checked through numerous applications (see, e.g.,Bargain (2006)).

11

units, ascertains which are eligible for that instrument and determines the amount

of bene�t or tax liability for each member of the unit. Finally, after all taxes and

bene�ts in question are simulated, disposable income is calculated.

3.2 Scenarios

The existing literature on stabilization so far has concentrated on increases in earn-

ings or gross incomes to examine the stabilizing impact of tax bene�t systems. In

the light of the current economic crisis, there is much more interest in a downturn

scenario. Reinhart and Rogo¤ (2009) stress that recessions which follow a �nancial

crisis have particularly severe e¤ects on asset prices, output and unemployment.

Therefore, we are interested not only in a scenario of a uniform decrease in in-

comes but also in an increase of the unemployment rate. We compare a scenario

where gross incomes are decreased by 5% (income shock) to a scenario where the

unemployment rate increases by �ve percentage points (unemployment shock).12

The increase of the unemployment rate is modeled through reweighting of our

samples.13 The weights of the unemployed are increased while those of the employed

with similar characteristics are decreased, i.e., in e¤ect, a fraction of employed house-

holds is made unemployed. With this reweighting approach we control for several

individual and household characteristics that determine the risk of becoming unem-

ployed (see Appendix A.2). The implicit assumption behind this approach is that

the socio-demographic characteristics of the unemployed remain constant.14

12Our scenarios can be seen as a conservative estimate of the expected impact of the currentcrisis (see Reinhart and Rogo¤ (2009) for e¤ects of previous crises). The (qualitative) results arerobust with respect to di¤erent sizes of the shocks. It would be further possible to derive morecomplicated scenarios with di¤erent shocks on di¤erent income sources or a combination of incomeand unemployment shock. However, this would only have an impact on the distribution of changeswhich are not relevant in the analysis of this paper. Therefore, we focus on these two simplescenarios in order to make our analysis as simple as possible. One should note, though, that ouranalysis is not a forecasting exercise. We do not aim at quantifying the exact e¤ects of the currenteconomic crisis but of stylized scenarios in order to explore the build-in automatic stabilizers ofexisting pre-crisis tax-bene�t systems. Conducting an ex-post analysis would include discretionarygovernment reactions and behavioral responses (see, e.g., Aaberge et al. (2000) for an empiricalex-post analysis of a previous crisis in the Nordic countries) and we would not be able to identifythe role of automatic stabilization.

13For the reweigthing procedure, we follow the approach of Immvervoll et al. (2006), who havealso simulated an increase in unemployment through reweighting of the sample. Their analysisfocuses on changes in absolute and relative poverty rates after changes in the income distributionand the employment rate.

14Cf. Deville and Särndal (1992) and DiNardo et al. (1996). This approach is equivalent toestimating probabilities of becoming unemployed (see, e.g., Bell and Blanch�ower (2009)) and then

12

4 Results

4.1 US vs. Europe

We start our analysis by comparing the US to Europe. Our simulation model in-

cludes 19 European countries which we treat as one single country (i.e. the �United

States of Europe�). All of them are EU member states, which is why we refer to

this group as the EU, bearing in mind that some EU member countries are missing.

We also consider the countries of the Euro area and refer to this group as �Euro�.

Figure 1 summarizes the results of our baseline simulation, which focuses on the

income tax, social insurance contributions (or payroll taxes) paid by employees and

bene�ts. Consider �rst the income shock. Approximately 38% of such a shock would

be absorbed by automatic stabilizers in the EU (and Euroland). For the US, we

�nd a slightly lower value of 32%. This di¤erence of just six percentage points is

surprising in so far as automatic stabilizers in Europe are usually considered to be

drastically higher than in the US.15 Our results qualify this view to a certain degree,

at least as far as proportional income shocks are concerned. Figure 1 shows that

taxes and social insurance contributions are the dominating factors which drive � in

case of a uniform income shock. Bene�ts are of minor importance in this scenario.

In the case of the unemployment shock, the di¤erence between the EU and the

US is larger. EU automatic stabilizers now absorb 48% of the shock (in the Euro

zone, we are close to 50%) whereas the stabilization e¤ect in the US is only 34%.

This di¤erence can be explained with the importance of unemployment bene�ts

which account for a large part of stabilization in Europe in this scenario. Table

7 in the Appendix shows that bene�ts alone absorb 19% of the shock in Europe

compared to just 7% in the US.

How does this cushioning of shocks translates into demand stabilization? The

results for stabilization of aggregate demand are shown in Figure 2. The demand

stabilization coe¢ cients are lower than the income stabilization coe¢ cients since

demand stabilization can only be achieved for liquidity constrained households.

selecting the individuals with the highest probabilities when controlling for the same characteristicsin the reweighting estimation (see Herault (2009)). The reweighting procedure is to some extentsensitive to changes in control variables. However, this mainly a¤ects the distribution of the shock(which we do not analyze) and not the overall or mean e¤ects which are important for the analysisin this paper.

15Note that for the US the value of the stabilization coe¢ cient for the federal income tax onlyis below 25% which is in line with the results of Auerbach and Feenberg (2000).

13

Figure 1: Decomposition of stabilization coe¢ cient for both scenarios

0

.1

.2

.3

.4

.5

.6

Inco

me

Stab

ilizat

ion

Coe

ffici

ent

EU EURO USAIncome Unemp. Income Unemp. Income Unemp.

FED Tax State TaxSIC Benefits

Source: Own calculations based on EUROMOD and TAXSIM

Therefore, the picture changes signi�cantly. For the EU, the cushioning e¤ect of

automatic stabilizers is now equal to 26%. For the US, we �nd a value of 19%, 7

percentage points less than for the EU. For the Euro area, where fewer households

are identi�ed to be credit constrained, the demand stabilization coe¢ cient (24%)

is lower than for the EU-group. For the unemployment shock, the picture again

changes completely. In the EU, the stabilization e¤ect is equal to 35%, the Euro

area is slightly lower (34%), whereas the value for the US (19%) is close to the value

for the income shock. These results suggest that the transfers to the unemployed, in

particular the rather generous systems of unemployment insurance in Europe, play

a key role for demand stabilization and drive the di¤erence in automatic stabilizers

between Europe and the US.

14

Figure 2: Stabilization coe¢ cient with and without liquidity constraints

0

.1

.2

.3

.4

.5

.6

Stab

ilizat

ion

Coe

ffici

ent T

AU (w

ithou

t VAT

)

EU EURO USAIncome Unemp. Income Unemp. Income Unemp.

Income Stabilization Demand Stabilization

Source: Own calculations based on EUROMOD and TAXSIM

4.2 Country decomposition

The results for the stabilization coe¢ cient vary considerably across countries, as

can be seen from Figure 3 (and Tables 6 and 7 in the Appendix). In the case of

the income shock, we �nd the highest stabilization coe¢ cient for Denmark, where

automatic stabilizers cushion 56% of the shock. Belgium (53%), Germany (48%)

and, surprisingly, Hungary (48%) also have strong automatic stabilizers. The lowest

values are found for Estonia (25%), Spain (28%) and Greece (29%). With the

exception of France, taxes seem to have a stronger stabilizing role than social security

contributions.

In case of the unemployment shock, the stabilization coe¢ cients are getting larger

in the majority of countries. Again, the highest value emerges for Denmark (71%),

followed by Sweden (69%), Austria (67%) Belgium (66%) and Germany (65%).

The relatively low value of stabilization from (unemployment) bene�ts in Finland

15

compared to its neighboring Nordic countries might be surprising at a �rst glance but

can be explained with the fact that Finland has the least generous unemployment

bene�ts of the Nordic countries (see Aaberge et al. (2000)). Hungary (46%) is now

below the EU average (48%) due to the very low level of unemployment bene�ts.

At the other end of the spectrum, there are some countries with values far below

the US level of 34%. These include Estonia (17%), Spain (28%), Poland (30%) and,

to a lesser extent, Italy (36%). The negative stabilization coe¢ cient for bene�ts in

Estonia and Poland can be explained with the fact that the majority of bene�ts is

conditional on working.

Figure 3: Decomposition of income stabilization coe¢ cient in both scenarios fordi¤erent countries

0.1.2.3.4.5.6.7.8

0.1.2.3.4.5.6.7.8

EESP

GRPL

PTSI

USAIT

UKIR

FRLU

EUEURO

FINL

SWAT

HUGE

BEDK

EESP

PLUSA

ITGR

SIIR

UKHU

NLEU

EUROLU

FIFR

PTGE

BEAT

SWDK

Income

Unemployment

Direct Tax SICBenefits

Inco

me

Stab

ilizat

ion

Coe

ffici

ent

Source: Own calculations based on EUROMOD and TAXSIM

When looking only at the personal income tax, it is surprising that the values

for the US (federal and state level tax) are higher than the EU average. This

quali�es to some extent the view that tax progressivity is higher in Europe (e.g.,

Alesina and Glaeser (2004) or Piketty and Saez (2007)). Of course, this can be

16

partly explained with the large heterogeneity within Europe. But still, only a few

countries like Belgium, Germany and the Nordic countries have higher contributions

of stabilization coming from the personal income tax.

How does this stabilization of disposable incomes a¤ect household demand? In

most Eastern European countries, households are more likely to be credit constrained

than in Western Europe because �nancial wealth is typically lower. Our estimates

con�rm this as can be seen from the �rst column of Table 1.16 For this reason auto-

matic stabilizers will be more important for demand stabilization in these countries.

This explains why we �nd the highest modi�ed demand stabilization coe¢ cient for

Hungary (46%) and why we �nd a stabilization e¤ect which is above or near to the

EU average even for Poland (30%) and Estonia (25%), although disposable income

stabilization is below the EU average in these countries. Relatively low values for

automatic stabilization e¤ects of the tax and transfer systems on demand are now

found in countries where households are relatively wealthy, so that credit constraints

are less important. These include Sweden, with a stabilization coe¢ cient of 26%,

Germany (25%) and in particular France (16%). Our results, including those for

other individual EU countries, are summarized in Table 1.17

4.3 Extension: Employer contributions and consumption

taxes

One objection to our results could be that we neglect some taxes which are poten-

tially relevant as automatic stabilizers. These include consumption taxes like the

value added tax or sales taxes, as well as social insurance contributions or payroll

taxes paid by employers, corporate income taxes and other taxes like e.g. property

taxes. The stabilization of cash �ows of corporations has implications for aggregate

demand which di¤er substantially from the implications of stabilizing household dis-

posable income and analyzing these implications would be beyond the scope of this

paper.18 In this section we present results for the income stabilization coe¢ cient

including employer and employee social insurance contributions and consumption

16As liquidity-constrained households are those households with low wealth and thus typicallylow income, one can expect that their share of income is lower than their share in the totalpopulation. In our data, this is true for all countries (see column 2 of Table 1).

17The results are robust to other de�nitions of liquidity constraints - at least with respect tocross-country rankings (see also Appendix Table 8). Of course, the higher the share of liquidityconstrained households the higher �C :

18This issue is discussed in Devereux and Fuest (forthcoming) and Buettner and Fuest (2009).

17

Table 1: Stabilization of aggregate demandShare liquidity constrained Income shock Unemployment shockPopulation Income � Income �C Demand � Income �C Demand

AT 0.891 0.883 0.439 0.388 0.670 0.606BE 0.706 0.639 0.527 0.348 0.657 0.466DK 0.619 0.576 0.558 0.320 0.707 0.470EE 0.984 0.969 0.253 0.246 0.168 0.162FI 0.741 0.629 0.396 0.266 0.541 0.385FR 0.479 0.420 0.370 0.161 0.582 0.341GE 0.511 0.503 0.481 0.248 0.645 0.376GR 0.854 0.822 0.291 0.234 0.383 0.310HU 0.976 0.961 0.476 0.457 0.464 0.452IR 0.736 0.646 0.363 0.228 0.425 0.315IT 0.762 0.733 0.346 0.283 0.359 0.268LU 0.773 0.768 0.374 0.284 0.533 0.419NL 0.706 0.657 0.397 0.264 0.472 0.348PL 0.985 0.982 0.301 0.296 0.295 0.288PT 0.899 0.886 0.303 0.273 0.625 0.589SI 0.743 0.632 0.317 0.112 0.425 0.245SP 0.837 0.824 0.277 0.225 0.283 0.230SW 0.630 0.599 0.420 0.257 0.685 0.501UK 0.824 0.775 0.352 0.277 0.441 0.397EU 0.717 0.686 0.378 0.258 0.482 0.350EURO 0.657 0.626 0.385 0.242 0.504 0.339USA 0.777 0.536 0.322 0.194 0.337 0.189

Source: Own calculations based on EUROMOD and TAXSIM.Note: A household is de�ned as liquidity constrained if its capitalized wealth is less than

the disposable income of at least two months (cf. Zeldes (1989)).

taxes.

Comparing Table 2 with Table 6 and Table 3 with Table 7, respectively, several

interesting results become apparent. Clearly, stabilization through the consumption

taxes strongly depends on their level. Thus, it is not surprising that high values for

the stabilization coe¢ cient of consumption taxes can be found for countries such as

Denmark, Estonia, Finland, Ireland, the Netherlands or Slovenia which all belong

to the group of countries with high levels of indirect taxation. In contrast, the

respective coe¢ cient for the US is at the lower end.

With regard to the new values of stabilization throug employer SIC, it becomes

evident that the gap between the EU and the US widens substantially. Comparing

18

results with and without employer social insurance contributions, we �nd for stabi-

lization through SIC that the EU-US gap has increased by roughly 10 percentage

points in both scenarios. As a consequence, overall stabilization, shown in the last

column of Tables 2 and 3 (and Tables 6 and 7 in the Appendix), is now signi�-

cantly higher in Europe compared with the US not only in the unemployment, but

also in the income shock scenario. Hence, expanding the set of taxes by employer

social insurance contributions and consumption taxes, we �nd an increasing gap in

automatic stabilization between the EU and the US for both shock scenarios.

Table 2: Decomposition income scenario with employer and employee SIC and CTFEDTax StateTax SIC BEN TaxSicBen CT TSBCT

AT 0.253 0.000 0.258 0.006 0.517 0.107 0.624BE 0.317 0.000 0.278 0.012 0.607 0.084 0.691DK 0.447 0.000 0.101 0.017 0.566 0.145 0.711EE 0.174 0.000 0.257 0.003 0.433 0.126 0.559FI 0.281 0.000 0.215 0.005 0.501 0.140 0.641FR 0.092 0.000 0.508 0.022 0.622 0.077 0.699GE 0.314 0.000 0.211 0.010 0.535 0.086 0.622GR 0.187 0.000 0.157 0.000 0.345 0.101 0.447HU 0.243 0.000 0.335 0.007 0.585 0.110 0.695IR 0.295 0.000 0.087 0.013 0.395 0.145 0.540IT 0.210 0.000 0.238 0.011 0.458 0.094 0.552LU 0.243 0.000 0.173 0.011 0.427 0.137 0.564NL 0.267 0.000 0.124 0.011 0.402 0.145 0.547PL 0.148 0.000 0.223 0.013 0.384 0.120 0.504PT 0.170 0.000 0.239 0.009 0.417 0.115 0.533SI 0.287 0.000 0.038 0.028 0.321 0.160 0.481SP 0.205 0.000 0.175 0.001 0.382 0.098 0.479SW 0.286 0.000 0.254 0.010 0.549 0.120 0.669UK 0.246 0.000 0.128 0.029 0.403 0.115 0.518EU 0.223 0.000 0.241 0.014 0.478 0.101 0.579EURO 0.222 0.000 0.265 0.011 0.497 0.095 0.592USA 0.240 0.049 0.077 -0.006 0.360 0.031 0.392

Source: Own calculations based on EUROMOD and TAXSIM

5 Discussion of the results

In this section, we discuss a number of possible objections to and questions raised

by our analysis. These include the relation of our results to widely used macro

19

Table 3: Decomposition unemployment scenario with employer and employee SICand CT

FEDTax StateTax SIC BEN TaxSicBen CT TSBCTAT 0.169 0.000 0.297 0.256 0.721 0.062 0.783BE 0.217 0.000 0.262 0.232 0.711 0.062 0.772DK 0.239 0.000 0.095 0.377 0.711 0.097 0.808EE 0.133 0.000 0.270 -0.024 0.379 0.138 0.517FI 0.183 0.000 0.223 0.218 0.625 0.106 0.730FR 0.047 0.000 0.501 0.195 0.743 0.052 0.795GE 0.202 0.000 0.253 0.234 0.690 0.058 0.747GR 0.118 0.000 0.195 0.111 0.424 0.089 0.514HU 0.179 0.000 0.361 0.037 0.577 0.112 0.689IR 0.198 0.000 0.077 0.174 0.449 0.132 0.581IT 0.150 0.000 0.262 0.062 0.475 0.091 0.565LU 0.137 0.000 0.157 0.274 0.567 0.103 0.671NL 0.093 0.000 0.215 0.216 0.523 0.115 0.639PL 0.130 0.000 0.285 -0.023 0.392 0.118 0.511PT 0.193 0.000 0.221 0.263 0.677 0.064 0.741SI 0.154 0.000 0.311 0.048 0.495 0.119 0.614SP 0.103 0.000 0.245 0.074 0.422 0.091 0.513SW 0.151 0.000 0.257 0.350 0.759 0.064 0.823UK 0.181 0.000 0.123 0.174 0.477 0.100 0.578EU 0.146 0.000 0.269 0.153 0.568 0.083 0.651EURO 0.138 0.000 0.296 0.166 0.600 0.074 0.674USA 0.174 0.041 0.102 0.071 0.388 0.030 0.418

Source: Own calculations based on EUROMOD and TAXSIM

indicators of automatic stabilizers, the role of other taxes, the correlation between

automatic stabilizers and other macro variables like e.g. openness and, �nally, the

correlations between discretionary �scal stimulus programs and automatic stabilizers

as well as openness.

5.1 Stabilization coe¢ cients and simple macro indicators

One could argue that aggregate measures like e.g. the tax revenue to GDP ratio re-

veal su¢ cient information on the magnitude of automatic stabilizers in the di¤erent

countries. For instance, the IMF (2009) has recently used aggregate tax to GDP

ratios as proxies for the size of automatic stabilizers in G-20 countries. The upper

panel of Figure 4 depicts the relation between the ratio of average revenue 2006-2010

20

to GDP and the income stabilization coe¢ cient for the proportional income shock.19

With a correlation of 0.58, one can conclude that government size is indeed a good

predictor for the amount of automatic stabilization. The picture changes, however,

if stabilization of aggregate household demand is considered, i.e. if we account for

liquidity constraints. As shown in Figure 4 (lower panel), with a coe¢ cient of 0.33

government size and stabilization of aggregate household demand are only weakly

correlated.20

Figure 4: Government size and income and demand stabilization coe¢ cients

AT

BEDK

EE

FIFR

GE

GR

HU

IR ITLU NL

PL PTSISP

SW

UKUSA

0

.1

.2

.3

.4

.5

.6

Inco

me

Sta

biliz

atio

n C

oeffi

cien

t

.3 .35 .4 .45 .5 .55 .6Average annual revenue to GDP, 2007-2010

ATBE

DK

EE FI

FR

GEGR

HU

IRITLU NL

PL PT

SI

SPSWUK

USA

0

.1

.2

.3

.4

.5

.6

Dem

and

Sta

biliz

atio

n C

oeffi

cien

t

.3 .35 .4 .45 .5 .55 .6Average annual revenue to GDP, 2007-2010

Source: Own calculations based on EUROMOD and TAXSIM, European Commission(2009a).

These simple correlations suggest that macro indicators like tax revenue to GDP

ratios are meaningful indicators for the stabilization e¤ect of the tax and transfer

19All �gures and correlations in this section are population-weighted in order to control fordi¤erent country sizes. However, results are similar to those without population-weighting. Wealso obtained similar results when using the government spending to GDP ratio instead of revenueas a measure of the size of the government.

20The respective correlations for the unemployment shock are 0.72 and 0.52.

21

system on disposable income but can be misleading as indicators of the stabilization

e¤ect on household demand. The reason is that the latter depends on the presence of

liquidity constraints. The income share of liquidity constrained households, however,

is negatively correlated with the size of government. In our analysis, we �nd a

correlation of -0.30 (see also Figure 7 in the Appendix).

Another interesting point arises from Figure 4 when making vertical comparisons

between similar countries. For instance, Denmark and Sweden, and - to some extent

- Belgium and France have similar levels of revenue to GDP ratios. However, the

stabilization is higher in Denmark and Belgium. In both countries, the importance

of the (progressive) income tax is higher, whereas Sweden and France rely more on

proportional social insurance contributions. Therefore, not only the size but also the

structure of the tax bene�t system are important for its possibilities of automatic

stabilization.

5.2 Automatic stabilizers and openness

It is a striking feature of our results that automatic stabilizers di¤er signi�cantly

within Europe. In particular, automatic stabilizers in Eastern and Southern Euro-

pean countries are much weaker than in the rest of Europe. One factor contributing

to this is that government size is often positively correlated with per capita incomes,

at least in Europe. The stabilization of disposable incomes will therefore be higher

in high income countries, just as a side e¤ect of a larger public sector.

But di¤erences in automatic stabilizers across countries may also have other rea-

sons. In particular, the e¤ectiveness of demand stabilization as a way of stabilizing

domestic output is smaller, the more open the economy. In very open economies, do-

mestic output will depend heavily on export demand and higher demand by domestic

households will partly lead to higher imports. Clearly, openness of the economy has

a number of other implications for the tax and transfer system, including the view

that more open economies need more insurance against shocks as argued, e.g., by

Rodrik (1998). Figure 5 depicts the relationship between income stabilization coef-

�cients and openness as measured by the ratio of exports plus imports over GDP.

As graph 5 shows, it is not the case that more open economies have weaker auto-

matic stabilizers, the correlation is even positive (0.57). Our results thus support

the hypothesis of Rodrik (1998) that income stabilization is higher in more open

economies. For the demand stabilization coe¢ cient, we �nd a similar correlation.

22

Figure 5: Income stabilization coe¢ cient and openness of the economy

AT

BEDK

EE

FIFR

GE

GR

HU

IRIT

LUNL

PLPT SI

SP

SW

UKUSA

0

.1

.2

.3

.4

.5

.6

Inco

me

stab

iliza

tion

coef

ficie

nt

0 .25 .5 .75 1 1.25 1.5 1.75 2 2.25 2.5 2.75Average ratio of exports and imports to GDP, 2000-2004

Income stabilization coefficient Fitted values

Source: Own calculations based on EUROMOD and TAXSIM, Heston et al. (2006).

5.3 Automatic stabilizers and discretionary �scal policy

In the debate on �scal policy responses to the crisis, some countries have been

criticized for being reluctant to enact �scal stimulus programs in order to stabilize

demand, in particular Germany. One reaction to this criticism was to point to the

fact that automatic stabilizers in Germany are more important than in other coun-

tries, so that less discretionary action is required. This raises the general question of

whether countries with weaker automatic stabilizers have taken more discretionary

�scal policy action. To shed some light on this issue, we relate the size of �scal

stimulus programs to stabilization coe¢ cients.

Graph 6 shows that income stabilization coe¢ cients are largely uncorrelated to

the size of �scal stimulus programs (-0.10). A larger negative correlation emerges

when we consider demand stabilization coe¢ cients (see Graph 8 in the Appendix).

Our �nding of a small correlation between automatic stabilizers and discretionary

23

Figure 6: Discretionary measures and income stabilization coe¢ cient

AT

BEDK

EE

FI

FR

GE

GRHU

IRIT

LU

NL

PL

PT

SI

SP

SW

UK

USA

0

.005

.01

.015

.02

.025

Aver

age

annu

al d

iscr

etio

nary

mea

sure

s 20

09-2

010

.2 .3 .4 .5 .6Income stabilization coefficient

Average annual discretionary measures 2009-2010 Fitted values

Source: Own calculations based on EUROMOD and TAXSIM, European Commission(2009c), IMF (2009) and International Labour O¢ ce and International Institute for

Labour Studies (2009).

measures quali�es the view that countries with lower automatic stabilizers have

engaged in more discretionary �scal policy action (e.g., IMF (2009), p. 27).

A further concern in the policy debate put forward by supporters of large and

coordinated discretionary measures is that countries could limit the size of their pro-

grams at the expense of countries with more generous �scal policy responses. The

idea behind this argument is that some countries might show a free-rider behavior

and pro�t from spill-over e¤ects of discretionary measures.21 Therefore, we inves-

tigate the hypothesis if more open countries which are supposed to bene�t more

from spill-over e¤ects indeed passed smaller stimulus programs. We �nd a negative

correlation of -0.40 between the average annual discretionary measures in 2009 and

2010 and the coe¢ cient for openness which supports the hypothesis.22

21In that sense, a �scal stimulus program can be seen as a positive externality since potentialpositive e¤ects are not limited to the country of origin.

22Cf. Graph 9 in the Appendix. A multivariate regression of discretionary measures on theincome stabilization coe¢ cients, a measure of openness of the respective economies and their

24

6 E¤ects on income distribution (preliminary)

After having analyzed the shock absorption capacities of the tax and transfer sys-

tems in Europe and the US, the logical next step is to consider the distributional

consequences of the shock scenarios. A precise analysis on distributional e¤ects of

the economic crisis of 2008-2009 has to wait until more recent data are available.

Nevertheless, it is interesting to note that the shock scenarios analyzed in this paper

lead to somewhat di¤erent implications for the income distribution as is shown in

Tables 4 and 5. Both tables report percentage changes in inequality measures (Gini

and the Generalized Entropy Measures 0-2) and in headcount measures for being

poor or rich.23 While the proportional income shock leads to reductions in all in-

equality measures, the overall picture for the asymmetric unemployment shock is not

that clear. In some countries, this shock scenario leads to an increase in inequality

indicating that distributional implications crucially depend on which income groups

are hardest hit by unemployment and income losses. If low income groups are the

�rst who loose their jobs during a recession, one can expect an increase in inequal-

ity. However, if also middle or upper income groups are a¤ected which seems to be

relevant especially in long-lasting recessions such as the current one, distributional

implications become more ambiguous. Therefore, future research will need to show

which distributional implications can be drawn from the current economic crisis.

governments�budget balance in 2007 leads to signi�cant coe¢ cients of openness and the budgetbalance; whereas the relationship between discretionary �scal policy and the amount of automaticstabilization remains insigni�cant. This result indicates that in addition to the argument aboveabout openness, some governments have been constrained by weak budget positions in their decisionmaking about discretionary �scal policy. However, due to the very small sample size, this inferenceshould be interpreted with caution.

23Note that a family is counted as poor if its equivalised disposable income is lower than 60%of the mean equivalised disposable income and as rich if its income exceeds twice the mean.

25

Table 4: E¤ect of proportional income shock on income distributionGini GE0 GE1 GE2 Poor Rich

at -1.297 -2.633 -2.352 -2.167 4.760 -12.088be -2.270 -4.023 -4.129 -4.603 2.673 -16.241dk -2.064 -3.063 -3.459 -3.851 3.838 -18.903ee -1.622 -3.734 -3.062 -3.184 4.529 -11.508� -1.806 -3.324 -3.066 -3.926 5.622 -13.981fr -1.422 -2.811 -2.492 -2.462 7.458 -9.947ge -1.489 -3.102 -2.821 -2.838 4.141 -12.982gr -0.338 -0.683 -0.819 -1.181 7.288 -11.355hu -0.604 -1.097 -1.675 -2.768 5.701 -9.241ir -1.335 -3.017 -2.639 -2.989 3.701 -12.591it -0.735 -1.076 -1.478 -1.840 4.910 -5.857lu -1.233 -2.590 -2.684 -3.112 9.994 -14.276nl -1.232 -2.507 -2.233 -2.084 10.629 -16.256pl -0.923 -1.917 -2.115 -2.671 6.749 -9.692pt -0.611 -1.578 -1.093 -1.013 4.693 -6.055si -0.318 -0.650 -0.892 -1.395 0.273 -1.290sp -0.693 -1.091 -1.261 -1.390 6.343 -13.806sw -2.050 -3.585 -3.305 -3.025 4.215 -15.446uk -2.219 -4.439 -3.912 -3.756 3.753 -13.001

Source: Own calculations based on EUROMOD and TAXSIM

26

Table 5: E¤ect of unemployment shock on income distributionGini GE0 GE1 GE2 Poor Rich

at 0.304 0.964 0.713 0.688 4.421 -3.619be 0.126 -0.258 -0.250 -0.230 3.869 -4.322dk -0.218 -0.505 1.515 6.731 1.176 -5.054ee 0.914 1.971 1.808 1.976 6.542 -2.989� 0.347 0.340 -0.355 -1.537 7.104 -3.428fr 0.210 0.371 0.428 0.479 4.083 -2.409ge 0.445 0.675 0.846 1.006 6.245 -3.469gr 0.166 0.168 0.352 0.519 2.509 -2.820hu 0.518 1.077 0.988 1.304 5.612 -3.861ir 1.154 1.566 0.995 -1.235 10.295 -7.285it 0.507 1.463 0.969 1.185 3.567 -2.234lu -0.225 -0.447 -0.152 0.186 1.335 -3.843nl 0.652 2.095 1.327 1.115 7.892 -3.985pl 0.281 0.690 0.290 0.010 3.757 -2.639pt -0.709 -1.435 -1.556 -1.912 1.528 -2.667si 0.327 0.583 0.503 0.429 4.354 -2.931sp 0.590 1.455 1.173 1.327 3.545 -3.003sw -0.037 -0.329 -0.213 -0.523 3.406 -3.177uk 1.074 2.053 2.119 2.386 7.895 -2.873

Source: Own calculations based on EUROMOD and TAXSIM

27

7 Conclusions

In this paper we have used the microsimulation models for the tax and transfer sys-

tems of 19 European countries (EUROMOD) and the US (TAXSIM) to investigate

the extent to which automatic stabilizers cushion household disposable income and

household demand in the event of macroeconomic shocks. Our analysis has �rst

focused on the personal income tax, employee social insurance contributions and

bene�ts, then an extended analysis including employer social insurance contribu-

tions and consumption taxes has been presented. One of the key �ndings of our

analysis is that the amount of automatic stabilization depends strongly on the type

of income shock. In the case of a proportional income shock, approximately 38%

of the shock would be absorbed by automatic stabilizers in the EU. For the US, we

�nd a value of 32%. Within the EU, there is considerable heterogeneity, and results

range from a value of 25% for Estonia to 56% for Denmark. In general automatic

stabilizers in Eastern and Southern European countries are considerably lower than

in Continental and Northern European countries.

In the case of an unemployment shock, which a¤ects households asymmetrically,

the di¤erence between the EU and the US is larger. EU automatic stabilizers absorb

48% of the shock whereas the stabilization e¤ect in the US is only 34%. Again, there

is considerable heterogeneity within the EU.

These results suggest that social transfers, in particular the rather generous sys-

tems of unemployment insurance in Europe, play a key role for the stabilization of

disposable incomes and household demand and explain a large part of the di¤erence

in automatic stabilizers between Europe and the US. This is con�rmed by the de-

composition of stabilization e¤ects in our analysis. In the case of the unemployment

shocks, bene�ts alone absorb 19% of the shock in Europe compared to just 7% in

the US, whereas the stabilizing e¤ect of income taxes (taking into account State

taxes in the US as well) is similar. This quali�es to some extent the view that the

tax wedge is larger in Europe than in the US. This is only true when looking at the

high tax countries like Belgium, Denmark, Finland, Germany or Sweden.

How does this cushioning of shocks translate into demand stabilization? Since

demand stabilization can only be achieved for liquidity constrained households, the

picture changes signi�cantly. For the income shock, the cushioning e¤ect of auto-

matic stabilizers is now equal to 26% in the EU. For the US, we �nd a value of 19%,

which is again rather similar. The value for the Euro area (24%) is close to the value

28

for the EU. For the unemployment shock, however, we �nd a large di¤erence. In

the EU, the stabilization e¤ect is equal to 35% whereas the value for the US (19%)

is close to the value for the income shock.

Does this mean that the US economy is particularly vulnerable to the current

economic crisis? To the extent that wages are more �exible than in Europe, one could

hope that fewer jobs will be lost in the crisis, so that the proportional income scenario

is more relevant. But US labor markets are also characterized by less employment

protection, so that job losses occur more quickly. For instance, between April 2008

and April 2009, the US unemployment rate increased by 3.9 percentage points while

the unemployment rate in the Euro area only increased by less than two percentage

points. There is much less automatic stabilization of disposable incomes as well as

household consumption demand than in Europe.

A second major result from our analysis is that demand stabilization di¤ers

considerably from disposable income stabilization. This has important policy im-

plications, also for discretionary �scal policy. As low income households are more

likely to be liquidity constrained and have a higher propensity of spending an in-

come increase, policies aimed at those households should lead to higher stabilizing

e¤ects. If liquidity constraints are low, reducing tax rates in order to tackle the

crisis will not be successful in increasing aggregate demand (see also Shapiro and

Slemrod (2009)). In this case, increasing government expenditure might be a more

successful way of tackling the crisis.

A third important result of our analysis is that automatic stabilizers are very

heterogenous within Europe. Interestingly, Eastern and Southern European coun-

tries are characterized by rather low automatic stabilizers. This is surprising, at

least from an insurance point of view because lower average income (and wealth)

implies that households are more vulnerable to income shocks. One explanation

for this �nding could be that countries with lower per capita incomes tend to have

smaller public sectors. From this perspective, weaker automatic stabilizers in East-

ern and Southern European countries are a potentially unintended side e¤ect of the

lower demand for government activity including redistribution. Another potential

explanation, the idea that more open economies have weaker automatic stabilizers

because domestic demand spills over to other countries, seems to be inconsistent with

the data, at least as far as the simple correlation between stabilization coe¢ cients

and trade to GDP ratios is concerned.

Finally, we have discussed the claim that countries with smaller automatic sta-

29

bilizers have engaged in more discretionary �scal policy action. According to our

results, there is no correlation between �scal stimulus programs of individual coun-

tries and stabilization coe¢ cients. However, we �nd that more open countries and

countries with higher budget de�cits have passed smaller stimulus programs. All

in all, our results suggest that policymakers did not take into account the forces

of automatic stabilizers when designing active �scal policy measures to tackle the

current economic crisis.

These results have to be interpreted in the light of various limitations of our

analysis. Firstly, the role of tax and transfer systems for stabilizing household

demand, not just disposable income, is based on strong assumptions on the link

between disposable income and household expenditures. Although we have used

what we believe to be the best available method for estimating liquidity constraints,

considerable uncertainty remains as to whether this method leads to an appropriate

description of household behavior. Secondly, our analysis abstracts from automatic

stabilization through other taxes, in particular corporate income taxes. Thirdly,

we have abstracted from the role of labor supply or other behavioral adjustments

for the impact of automatic stabilizers. We intend to pursue these issues in future

research.

A Appendix:

A.1 Additional results

30

Table 6: Decomposition income scenarioFEDTax StateTax SIC BEN TaxSicBen

AT 0.294 0.000 0.139 0.006 0.439BE 0.382 0.000 0.131 0.014 0.527DK 0.455 0.000 0.086 0.018 0.558EE 0.228 0.000 0.021 0.004 0.253FI 0.340 0.000 0.050 0.006 0.396FR 0.153 0.000 0.181 0.036 0.370GE 0.351 0.000 0.118 0.012 0.481GR 0.203 0.000 0.088 0.000 0.291HU 0.307 0.000 0.160 0.009 0.476IR 0.310 0.000 0.039 0.014 0.363IT 0.254 0.000 0.079 0.013 0.346LU 0.265 0.000 0.097 0.012 0.374NL 0.270 0.000 0.116 0.011 0.397PL 0.168 0.000 0.118 0.015 0.301PT 0.203 0.000 0.090 0.010 0.303SI 0.289 0.000 0.031 0.028 0.317SP 0.240 0.000 0.035 0.001 0.277SW 0.368 0.000 0.040 0.012 0.420UK 0.267 0.000 0.054 0.031 0.352EU 0.260 0.000 0.100 0.017 0.378EURO 0.263 0.000 0.108 0.015 0.385USA 0.240 0.049 0.039 -0.006 0.322Source: Own calculations based on EUROMOD and TAXSIM

31

Table 7: Decomposition unemployment scenarioFEDTax StateTax SIC BEN TaxSicBen

AT 0.200 0.000 0.167 0.303 0.670BE 0.257 0.000 0.124 0.276 0.657DK 0.243 0.000 0.083 0.382 0.707EE 0.178 0.000 0.022 -0.032 0.168FI 0.224 0.000 0.050 0.267 0.541FR 0.076 0.000 0.190 0.317 0.582GE 0.231 0.000 0.145 0.268 0.645GR 0.126 0.000 0.137 0.119 0.383HU 0.227 0.000 0.190 0.047 0.464IR 0.207 0.000 0.036 0.182 0.425IT 0.183 0.000 0.101 0.076 0.359LU 0.147 0.000 0.090 0.296 0.533NL 0.103 0.000 0.131 0.239 0.472PL 0.151 0.000 0.170 -0.027 0.295PT 0.225 0.000 0.094 0.306 0.625SI 0.175 0.000 0.216 0.054 0.425SP 0.127 0.000 0.064 0.091 0.283SW 0.197 0.000 0.029 0.458 0.685UK 0.194 0.000 0.061 0.186 0.441EU 0.172 0.000 0.121 0.189 0.482EURO 0.166 0.000 0.129 0.210 0.504USA 0.174 0.041 0.051 0.071 0.337Source: Own calculations based on EUROMOD and TAXSIM

32

Table 8: Stabilization of aggregate demandShare liquidity constrained Income shock Unemployment shockPopulation Income � Demand � Income � Demand � Income

AT 0.750 0.408 0.172 0.439 0.396 0.670BE 0.750 0.352 0.190 0.527 0.368 0.657DK 0.750 0.387 0.205 0.558 0.392 0.707EE 0.750 0.350 0.087 0.253 0.073 0.168FI 0.750 0.364 0.136 0.396 0.314 0.541FR 0.750 0.379 0.143 0.370 0.383 0.582GE 0.750 0.374 0.172 0.481 0.366 0.645GR 0.750 0.355 0.066 0.291 0.156 0.383HU 0.750 0.322 0.120 0.476 0.151 0.464IR 0.750 0.409 0.130 0.363 0.260 0.425IT 0.750 0.372 0.116 0.346 0.139 0.359LU 0.750 0.413 0.131 0.374 0.345 0.533NL 0.750 0.467 0.168 0.397 0.283 0.472PL 0.750 0.328 0.099 0.301 0.093 0.295PT 0.750 0.371 0.083 0.303 0.216 0.625SI 0.750 0.355 0.077 0.317 0.154 0.425SP 0.750 0.408 0.091 0.277 0.147 0.283SW 0.750 0.400 0.162 0.420 0.456 0.685UK 0.750 0.388 0.133 0.352 0.260 0.441EU 0.750 0.378 0.133 0.378 0.260 0.482EURO 0.750 0.384 0.136 0.385 0.277 0.504USA 0.750 0.310 0.089 0.322 0.128 0.337Source: Own calculations based on EUROMOD and TAXSIM. Note: Bottom 75% of

gross income distribution are assumed to be liquidity constrained.

33

Figure 7: Income share of liquidity constrained households and government revenue

AT

BEDK

EE

FI

FR

GE

GR

HU

IR

ITLU

NL

PL

PT

SI

SP

SW

UK

USA

0

.1

.2

.3

.4

.5

.6

.7

.8

.9

1S

hare

of i

ncom

e liq

uidi

ty c

onst

rain

ed h

ouse

hold

s

.3 .35 .4 .45 .5 .55 .6Average annual revenue 2007-2010 to GDP

Share of income liquidity constrained households Fitted values

Source: Own calculations based on EUROMOD and TAXSIM, European Commission(2009a).

Figure 8: Discretionary measures and demand stabilization

AT

BEDK

EE

FI

FR

GE

GRHU

IRIT

LU

NL

PL

PT

SI

SP

SW

UK

USA

0

.005

.01

.015

.02

.025

Aver

age

annu

al d

iscre

tiona

ry m

easu

res

2009

-201

0

0 .1 .2 .3 .4 .5 .6Demand stabilization coef ficient

Average annual discretionary measures 2009-2010 Fitted values

Source: Own calculations based on EUROMOD and TAXSIM, European Commission(2009c), IMF (2009).

34

Figure 9: Discretionary measures and openness of the economy

AT

BEDK

EE

FI

FR

GE

GRHU

IRIT

LU

NL

PL

PT

SI

SP

SW

UK

USA

0

.005

.01

.015

.02

.025

Ave

rage

ann

ual d

iscre

tiona

ry m

easu

res

2009

-201

0

.2 .7 1.2 1.7 2.2 2.7Average ratio of exports and imports to GDP, 2000-2004

Average annual discretionary measures 2009-2010 Fitted values

Source: Heston et al. (2006), European Commission (2009c), International Labour O¢ ceand International Institute for Labour Studies (2009) and IMF (2009).

35

A.2 Reweighting procedure for increasing unemployment

In order to increase the unemployment rate while keeping the aggregate counts of

other key individual and household characteristics constant, we follow the approach

taken by Immvervoll et al. (2006). The increase of the unemployment rates is mod-

eled through reweighting of our samples while controlling for several individual and

household characteristics that determine the risk of becoming unemployed.

We follow Immvervoll et al. (2006) and de�ne the unemployed as people aged

19�59 declaring themselves to be out of work and looking for a job. The within-

database national �unemployment rate�is calculated as the ratio of these unemployed

to those in the labor force, de�ned as the unemployed plus people aged 19�59 who

are (self)employed. The increased total number of unemployed people is calculated

by adding 5 percentage points to the �unemployment rate�within each country.

In EUROMOD, the baseline household weights supplied with the national data-

bases have been calculated to adjust for sample design and/or di¤erential non-

response (see Sutherland (2001) for details). Weights are then recalculated using the

existing weights as a starting point, but (a) using the increased (decreased) num-

ber of unemployed (employed) people as the control totals for them, and (b) also

controlling for individual demographic and household composition variables using

the existing grossed-up totals for these categories as control totals. The speci�c

variables used as controls are:

� employment status

� age (0�18, 19�24, 25�49, 50�59, 60+)

� gender

� marital status and household size

� education

� region

This method implies that the households without any unemployed people that

are similar to households with unemployed people (according to the above variables)

will have their weights reduced. In other words, these are the households who are

�made unemployed�in our exercise.

36

References

Aaberge, R., Björklund, A., Jäntti, M., Pedersen, P. J., Smith, N. and Wennemo,

T. (2000). Unemployment Shocks and Income Distribution: How Did the

Nordic Countries Fare during Their Crises?, Scandinavian Journal of Eco-

nomics 102(1): 77�99.

Alesina, A. and Glaeser, E. (2004). Fighting Poverty in the U.S. and Europe: A

World of Di¤erence, Oxford University Press.

Alesina, A., Glaeser, E. and Sacerdote, B. (2005). Work and leisure in the U.S. and

Europe: Why so di¤erent?, NBER Macroeconomcis Annual pp. 1�64.

Ankrom, J. (1993). An Analysis of Horizontal and Vertical Equity in Sweden, the

U.S. and the U.K, Scandinavian Journal of Economics 95: 119�24.

Auerbach, A. (2009). Implementing the New Fiscal Policy Activism, American

Economic Review: Papers and Proceedings 99:2: 543�549.

Auerbach, A. and Feenberg, D. (2000). The signi�cance of federal taxes as automatic

stabilizers, Journal of Economic Perspectives 14: 37�56.

Auerbach, A. and Hassett, K. (2002). Fiscal Policy and Uncertainty, International

Finance 5: 229�249.

Bargain, O. (2006). Microsimulation in action: policy analysis in Europe using

EUROMOD, vol. 25 of the series Research in Labor Economics, Elsevier.

Bayoumi, T. and Masson, P. (1995). Fiscal �ows in the United States and Canada:

Lessons for monetary union in Europe, European Economic Review 39: 253�274.

Bell, D. and Blanch�ower, D. (2009). What Should Be Done about Rising Unem-

ployment in the UK, IZA DP No. 4040.

Blanchard, O. (2006). Comments on �The Case Against the Case Against Discre-

tionary Policy�, by Alan Blinder, in R. Kopcke, G. Tootell and R. Triest (eds),

The macroeconomics of �scal policy, MIT Press, Boston, pp. 62�67.

37

Blanchard, O. and Perotti, R. (2002). An empirical characterization of the dynamic

e¤ects of changes in government spending and taxes on output, Quarterly Jour-

nal of Economics 117: 1329�1368.

Bourguignon, F. and Spadaro, A. (2006). Microsimulation as a tool for evaluating

redistribution policies, Journal of Economic Inequality 4(1): 77�106.

Buettner, T. and Fuest, C. (2009). The Role of the Corporate Income Tax as an

Automatic Stabilizer, Working Paper, Oxford University Centre for Business

Taxation.

Cameron, D. (1978). The Expansion of the Public Economy: A Comparative Analy-

sis, American Political Science Review 72: 1243�1261.

Devereux, M. and Fuest, C. (forthcoming). Is the Corporation Tax an E¤ective

Automatic Stabilizer?, National Tax Journal .

Deville, J.-F. and Särndal, C.-E. (1992). Calibration estimators in survey sampling,

Journal of the American Statistical Association 87: 376�382.

DiNardo, J., Fortin, N. and Lemieux, T. (1996). Labor Market Institutions and the

Distribution of Wages, 1973-1992: A Semiparametric Approach, Econometrica

64: 1001�1044.

Eaton, J. and Rosen, H. (1980). Optimal redistributive taxation and uncertainty,

Quarterly Journal of Economics 95: 357�364.

European Commission (2009a). Public Finances in EMU 2009. 10th edition.

European Commission (2009b). Taxation trends in the European Union - 2009

edition.

European Commission (2009c). The EU�s response to support the real economy

during the economic crisis: an overview of Member States�recovery measures.

Fatàs, A. and Mihov, I. (2001). Government size and automatic stabilizers, Journal

of International Economics 55: 3�28.

Feenberg, D. R. and Coutts, E. (1993). An Introduction to the TAXSIM Model,,

Journal of Policy Analysis and Management 12(1): 189�194.

38

Galí, J. (1994). Government size and macroeconomic stability, European Economic

Review 38: 117�132.

Girouard, N. and André, C. (2005). Measuring cyclically-adjusted budget balances

for OECD countries, OECD Economics Department Working Papers.

Goode, R. (1976). The Individual Income Tax, revised edition, Brookings., Wash-

ington.

Herault, N. (2009). Sequential Linking of Computable General Equilibrium and Mi-

crosimulation Models, WP No. 2/09, Melbourne Institute of Applied Economic

and Social Research, The University of Melbourne.

Heston, A., Summers, R. and Aten, B. (2006). Penn World Table Version 6.2,

Center for International Comparisons of Production, Income and Prices at the

University of Pennsylvania.

IMF (2009). Group of Twenty - Global Economic Policies and Prospects, Note by

the Sta¤ of the International Monetary Fund.

Immervoll, H., Kleven, H., Kreiner, C. and Saez, E. (2007). Welfare Reform in

European Countries: A Micro-Simulation Analysis, The Economic Journal 117(516): 1�44.

Immvervoll, H. (2004). Average and marginal e¤ective tax rates facing workers

in the EU. A micro-level analysis of levels, distributions and driving factors,

EUROMOD Working Paper Series.

Immvervoll, H., Levy, H., Lietz, C., Mantovani, D. and Sutherland, H. (2006).

The sensitivity of poverty rates to macro-level changes in the European Union,

Cambridge Journal of Economics 30: 181�199.

International Labour O¢ ce and International Institute for Labour Studies (2009).

The �nancial and economic crisis: A decent work response.

Johnson, D., Parker, J. and Souleles, N. (2006). Household Expenditure and the

Income Tax Rebates of 2001, American Economic Review 96(5): 1589�1610.

Kleven, H. J. and Kreiner, C. (2006). The Marginal Cost of Public Funds in OECD

Countries: Hours of Work versus Labor Force Participation, Journal of Public

Economics 90: 1955�1973.

39

Mabbett, D. (2004). Fiscal Stabilisers In Europe: The Macroeconomic Impact Of

Tax and Bene�t Systems, EUROMOD Working Paper No. EM7/04.

Mabbett, D. and Schelkle, W. (2007). Bringing macroeconomics back into the po-

litical economy of reform: the Lisbon Agenda and the ��scal philosophy�of the

EU, Journal of Common Market Studies 45: 81�104.

McIntyre, R. S., Denk, R., Francis, N., Gardner, M., Gomaa, W., Hsu, F. and Sims,

R. (2003). Who pays? A Distributional Analysis of the Tax Systems in All 50

States, Institute on Taxation and Economic Policy.

Mélitz, J. and Zumer, F. (2002). Regional redistribution and stabilization by the

central governement in canada; france; the uk and the us: A reassessment and

new tests, Journal of Public Economics 86 (2): 263�286.

Parker, J. (1999). The Reaction of Household Consumption to Predictable Changes

in Social Security Taxes, American Economic Review 89 (4): 959�973.

Pechman, J. (1973). Responsiveness of the Federal Income Tax to Changes in In-

come, Brookings Papers on Economic Activitiy 2: 385�421.

Pechman, J. (1987). Federal Tax Policy, Brookings, Washington.

Piketty, T. and Saez, E. (2007). How Progressive is the U.S. Federal Tax System?

A Historical and International Perspective, Journal of Economic Perspectives

21 (1): 3�24.

Prescott, E. C. (2004). Why do Americans work so much more than Europeans?,

Federal Reserve Bank of Minneapolis Quarterly Review 28 (1): 2�13.

Reinhart, C. and Rogo¤, K. (2009). The Aftermath of Financial Crisis, American

Economic Review: Papers & Proceedings 99 (2): 466�472.

Rodrik, D. (1998). Why Do More Open Economies Have Bigger Governments?,

Journal of Political Economy 106: 997�1032.

Sachs, J. and Sala-i Martin, X. (1992). Fiscal Federalism and Optimum Currency

Areas: Evidence for Europe from the United States, in M. B. Canzoneri, V.

Grilli and P. R. Masson editors, "Establishing a Central Bank: Issues in Europe

and Lessons from the U.S.", Cambridge University Press.

40

Saez, E. (2001). Using elasticities to derive optimal income tax rates, Review of

Economic Studies 68: 205�229.

Saez, E. (2002). Optimal income transfer programs: intensive versus extensive labor

supply responses, Quarterly Journal of Economics 117: 1039�1073.

Shapiro, M. D. and Slemrod, J. (1995). Consumer response to the timing of in-

come: Evidence from a change in tax withholding, American Economic Review

85(1): 274�283.

Shapiro, M. D. and Slemrod, J. (2003). Consumer response to tax rebates, American

Economic Review 93(1): 381�396.

Shapiro, M. D. and Slemrod, J. (2009). Did the 2008 tax rebates stimulate spending?,

American Economic Review: Papers & Proceedings 99(2): 374�379.

Souleles, N. (1999). The Response of Household Consumption to Income Tax Re-

funds, American Economic Review 89 (4): 947�958.

Sutherland, H. (2001). Final Report - EUROMOD: An Integrated European Bene�t-

Tax Model, EUROMOD Working Paper No. EM9/01.

Sutherland, H. (2007). Euromod: the tax-bene�t microsimulation model for the

European Union, in A. Gupta and A. Harding (eds), Modelling Our Future:

Population Ageing, Health and Aged Care, Vol. 16 of International Symposia in

Economic Theory and Econometrics, Elsevier, pp. 483�488.

van den Noord, P. (2000). The size and role of automatic stabilisers in the 1990s

and beyond, OECD Economics Department Working Papers, No.230.

Zeldes, S. (1989). Consumption and Liquidity Constraints: An Empirical Investiga-

tion, Journal of Political Economy 97 (2): 305�346.

41